[Crash-utility] Is there any support for userspace stack dump for crash utility?

[Dave Anderson]

----- Original Message -----
> Hi,
> 
> I'd like to find help here that does anyone have tried to get the
> crash utility support the userspace stack dumping, or it already be
> supported?
> Certainly userspace crash would not cause kernel panic, but for more
> robust product, we force kernel panic and trigger the kdump at every
> userspace crash.
> So the support for dumping user stack is import to us.
> 
> Anyone has some idea to this feature?
> If we could get the stack like stack dump from userspace to kernel
> space with parameter showing, then that is perfect.
> 
> Example:
> -010|do_vfs_ioctl(filp = 0xCE575500, ?, ?, arg = 1181448720)
> -011|sys_ioctl(fd = 27, cmd = 30000, arg = 1181448720)
> -012|ret_fast_syscall(asm)
> -->|exception
> -013|__ioctl(asm)
> -014|ioctl(?, request = 0)
> 
> Thanks,
> Lei

There's no chance that the base crash utility will ever do that.

However, in your case, you might try the "gcore" extension
module.  See its description, and download it from:

  http://people.redhat.com/anderson/extensions

It's help page (accessible from the page above):

  http://people.redhat.com/anderson/extensions/gcore_help_gcore.html

After you have gathered the vmcore, run the crash utility
on it, and use the "extend" command to load the "gcore" 
command.  Then use the "gcore" command to generate a user core 
dump file:

  crash> extend gcore.so
  gcore.so: shared object loaded
  crash> gcore <pid>
  Saved core.<pid>
  crash>

which will generate a core dump of that <pid>'s task.
Then you can run "gdb <command> core-<pid>".  If it
works as advertised, then you should be able to get
a backtrace up until the point it entered the kernel.

Dave